2016 年 56 巻 8 号 p. 1378-1385
A new cellular automaton (CA) model with a triple-layered multi-grid (TLMG) method for ternary alloys was developed, and prediction of microsegregation in Fe–C–Cr ternary alloys was performed by using the technique combining a two-dimensional CA model with the TLMG method and thermodynamic equilibrium (EQ) calculations. In this work, simulations of multi-dendritic growth were performed for twelve different Fe–C–Cr alloys by using the current CA model; then EQ calculations were conducted by utilizing the results of the previous CA simulations. The compositions of interdendritic liquids obtained by the CA simulations were used as initial input data for the EQ calculations. Microsegregation in the Fe–C–Cr alloys was evaluated using segregation indices corresponding to the ratios between the maximum and minimum concentrations in the calculation domains. The predicted values of a Cr segregation index increase with an increase in the carbon content in alloys; these values are in good agreement with the experimental data reported by Kishitake et al. [K. Kishitake, S. Matsuo and T. Okamoto Tetsu-to-Hagane, 63 (1977), p. 1108] The obtained results indicate that microsegregation in ternary alloys can be predicted quantitatively using the proposed method that includes CA simulations combined with EQ calculations.